![]() ![]() Stg.stage_twitter_sentiment: incremental stage table per tweetĭwh.fact_twitter_sentiment: historical fact table per tweetĭwh.dim_twitter_sentiment_afinn: dimension table for afinn score per wordĭwh.dim_twitter_sentiment_bing: dimension table for bing score per wordĭwh.view_twitter_sentiment: view that unions all fact and dim tablesįor Tableau analysis, I created two dashboards: Sentiment Analysis and Sentiment Comparison. Ods.twitter_sentiment: incremental load table per tweetĭs.ds_twitter_sentiment: historical table of ods per tweet The tables created and populated by the Alteryx workflow are: The schemas for this database are: ods, ds, stg, dwh. Fill in the application details and hit “create your Twitter application.”įor this project, a local SQL Server instance ((LocalDb)\TwitterSentiment) was set up with a local database (DB_Twitter_Sentiment). Next navigate to to create the API app (the one used for this project is shown below). The first step is to create a Twitter account (preferably just for this project). In order to extract the Twitter feed data, you need to create a Twitter API. Tableau: used to visualize and analyze the sentiment of the Twitter data. Microsoft SQL Server: used to store and host the data. R: used within Alteryx to perform sentiment analysis on the Twitter data. Twitter API: used to create an app that Alteryx can extract the data from.Īlteryx: used to extract and transform the data (including performing sentiment analysis). The tools and services that were used in this project are: We could then use this analysis to study and react to the sentiment of Twitter users who are tweeting about these data tools. The purpose of this project was to create a parameterizable process that could extract Twitter feed data about any business intelligence or ETL tool and perform sentiment analysis on that data. The solution here is to extract all of these tweets, analyze them to find those sentiments, and then have a way to visually explore them and isolate the insights. How many tweets referred to us in the last two weeks?Īnswering these types of questions can produce serious value for a business. It can help businesses answer questions like:ĭid that last marketing campaign we launched have any effect on how our newest product is viewed? This is where a sentiment analysis comes into play. ![]() Some companies utilize this as a source of customer thoughts and opinions, but it mostly remains an untapped mine of insight. Roughly 500 million tweets are posted by people every day, which translates to a rate of 6,000 tweets per second. ![]() The benefits were twofold: I could dabble with data science concepts, and also gain some insight into how some of the tools compare to one another on Twitter. I decided I would extract Twitter feed data about any business intelligence or ETL tool and perform a sentiment analysis on that data. I wanted to see what the buzz was about, and how I could increase my skill set with some new tools. As a consultant, I had never done anything related to data science before. At Keyrus, we have the opportunity to engage in “offline tasks.” An offline task allows you to work on a project that incorporates skills or technologies you don’t use in your day to day work. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |